Wine secondary aroma: understanding yeast production of higher alcohols

Contribution of yeasts to the sensory attributes of fermented foods and beverages goes far beyond sugar consumption and ethanol and carbon dioxide production. It includes some major by-products of fermentation, like glycerol or acetic acid, and hundreds of aroma-active compounds, including higher alcohols, esters, aldehydes, organic acids, volatile fatty acids or carbonyl compounds, as main constituents of the secondary or fermentation aroma of grape wine (Styger et al., 2011). In addition, they contribute to the enzymatic transformation of some neutral precursors, originating from the substrate, into odour-active molecules, so enhancing primary or varietal aroma in wine. The structural relationship between some higher alcohols, esters, organic acids and amino acids (which contribute some of the nitrogen sources in natural fermentation substrates) suggests the later might be the precursors of some important constituents of the secondary aroma. For example, isobutyl alcohol, active amyl alcohol and isoamyl alcohol are structurally related to valine, isoleucine and leucine respectively (Lambrechts and Pretorius, 2000), and they are linked through the Ehrlich pathway (Hazelwood et al., 2008). Although a contribution of amino acid carbon chains to the production of these higher alcohols has been clearly established (Reazin et al., 1970, 1973), there are also clear-cut evidences that the input from yeast central carbon metabolism is by no means negligible, being in most cases the main origin of these branched-chain alcohols (Rankine, 1967; Reazin et al., 1970). The topic of the dependence of higher alcohol production (and the relative contribution of each metabolic pathway) on the nitrogen content of the fermentation substrate has generated a wealth of scientific literature. However, drawing general rules has proven difficult, with results from different studies pointing in opposite directions. This suggests complex non-linear relationships between the availability of amino acids or nitrogen sources in general, and the release of aroma compounds by Saccharomyces cerevisiae. An illustration of complex dependences between aroma-active compounds and amino acid availability was shown by Hern andez-Orte et al. (2002). One intuitive idea about amino acid metabolism during fermentation, and more specifically in winemaking, is that, once uptaken by yeast, amino acids whose availability is below biosynthetic requirements, will be directed to protein synthesis (rather than being used for other purposes). Indeed, in order to avoid ‘in silico’ futile cycles, this assumption has been used in metabolic flux analysis in order to choose between anabolic or catabolic reactions to be incorporated in the metabolic network for each amino acid (Quir os et al., 2013). Early this year researchers from INRA at Montpellier published an elegant work based on stable isotope tracers, with an exhaustive analysis of the fate of amino acids in laboratory fermentation trials (Cr epin et al., 2017). According to their results, only a small proportion of the amino acids from synthetic must is directly incorporated into proteins by S. cerevisiae. Most of them are broken down by transamination and the amino groups used for de novo synthesis of proteinogenic amino acids. The level of transamination is roughly independent of amino acid availability or anabolic requirements. In agreement with other authors, and despite initial breakdown of amino acids is in the origin of the release of some higher alcohols, this is a minor contribution to the total amount finally produced. Central carbon metabolism plays a key role in the formation of these metabolites. However, not all of them strictly follow the above described general rules (Cr epin et al., 2017). In this issue, Rollero et al. (2017, this issue) used C-labelled leucine and valine, and introduce variations Received 5 June, 2017; accepted 14 June, 2017. *For correspondence. E-mail [email protected]; Tel. +34 941894980; Fax +34 941899728. Microbial Biotechnology (2017) 10(6), 1449–1450 doi:10.1111/1751-7915.12770 Funding information No funding information provided.